readable AI benchmarks (simplified)
Microsoft
Phi-3 Mini Instruct 3.8B
Phi-3 Mini Instruct 3.8B does not yet have enough benchmark data for a Quality Score. It has a 4,096 token context window.
Quality Scorenot availableQuality unavailable: missing AA-Omniscience outcomes
Value Scorenot available
Factual reliabilitynot available
Cache Discountnot available
Model specification
- Reasoning
- non-reasoning
- Input modalities
- none
- Output modalities
- none
- Context window
- 4,096 tokens
- Weights
- open weights
Token prices USD per 1M tokens
- Input
- not available
- Output
- not available
- Cached Input
- not available
- Cache write
- not available
- Cost per task
- not available
Capability
- Intelligence
- 5.8
- Coding
- not available
- Agentic
- not available
- Omniscience
- not available
- Correct
- not available
- Blended price
- not available
Answer outcomes
Fully graded outcomes (Correct + Incorrect): not available
Omniscience outcome data not available for this model.
Artificial Analysis benchmarks
- Humanity’s Last Exam
- 5%
- GPQA Diamond
- 32%
Similar models
- Phi-4 Multimodal InstructMicrosoftQuality Score not availableValue Score not available
- Phi-4MicrosoftQuality Score 15.6Value Score not available
- Phi-4 Mini InstructMicrosoftQuality Score 14.6Value Score not available
- NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIAQuality Score 11.4Value Score not available
- Gemma 3n E4B Instruct Preview (May '25)GoogleQuality Score not availableValue Score not available
- Gemini 1.0 UltraGoogleQuality Score not availableValue Score not available